This proposal is based on two years of preliminary work and is a direct and specific response to a request for applications aimed at generating animal models of cystic fibrosis (CF). Specific goals of the project will be to prepare null mutations and to transmit these to the mouse germline. Depending on the phenotype obtained with a null mutation, additional missense mutations and hypomorphic mutations will be generated. The various mutations would be introduced on to different mouse backgrounds in order to examine the effect of genetic variation at other loci on the phenotype in the mouse. The CF mutations would be moved on to other mouse strains either by backcrossing or by preparing ES cell lines from different mouse strains. The phenotype of the mutant mice would be assessed clinically, pathologically, and physiologically. The investigators already have committed at least two years work towards the goal of obtaining mice with mutations in the CF gene. Numerous constructs have been prepared in the region corresponding to human exon 19 and in the region corresponding to human exons 2 and 3. Insertion and replacement vectors are being investigated. Recombinant clones are being screened using a mini-Southern blot protocol and using PCR to analyze pools of clones. Clones documented to represent homologous recombinants by Southern blotting will be injected into mouse blastocysts. Animals with a high percent chimerism will be bred in order to obtain germline transmission. Homozygous mutant animals will be assessed for survival, weight gain, susceptibility to infection, and other phenotypic features known to be associated with CF. Complete pathological studies will be performed on mutant animals. Any residual function of the CF gene will be assessed using reverse transcriptase-PCR, Western blotting, and immunostaining of tissue sections. Animals will be provided to collaborators for electrophysiological analysis. Immortalized cell lines will be developed from mutant and wild-type animals using a variety of strategies. As indicated by this request for applications, there is substantial reason to believe that mouse models for CF would be valuable in developing a better understanding of the pathophysiology, in assessing pharmacologic approaches, and in developing somatic gene therapy.